Trace Element and Stable Isotope Geochemistry of Lwamondo and Zebediela Kaolins, Limpopo Province, South Africa: Implication for Paleoenvironmental Reconstruction

minerals

Article Trace Element and Stable Isotope Geochemistry of Lwamondo and Zebediela Kaolins, Limpopo Province, South Africa: Implication for Paleoenvironmental Reconstruction

Avhatakali Raphalalani 1, Georges-Ivo Ekosse 2,*, John Odiyo 1, Jason Ogola 1 and Nenita Bukalo 1

1 School of Environmental Sciences, University of Venda, Thohoyandou 0950, Limpopo Province, South Africa; [email protected] (A.R.); [email protected] (J.O.); [email protected] (J.O.); [email protected] (N.B.) 2 Directorate of Research and Innovation, University of Venda, Thohoyandou 0950, Limpopo Province, South Africa * Correspondence: [email protected]; Tel.: +27-159-628-504

Received: 6 December 2018; Accepted: 31 January 2019; Published: 4 February 2019

Abstract: The aim of the present study was the paleoenvironmental reconstruction of the prevailing environment under which the Lwamondo and Zebediela kaolin deposits were formed. Hence, this study reports deuterium and oxygen stable isotope values and trace and rare earth element concentrations for two samples of kaolin. Upper continental crust-normalised trace-element patterns reveal that large ion lithophile elements and high-ﬁeld-strength elements are generally depleted in Lwamondo and Zebediela kaolins, whereas transition trace elements are generally enriched in these kaolins. Upper continental crust-normalised rare earth element (REE) patterns show that there is a slight enrichment of heavy REEs (HREEs) compared to light REEs (LREEs) in these kaolins. The δ18O and δD stable isotope values for kaolinite from Lwamondo ranged from 17.4 to 19.1 and from −54 to 84 , respectively, whereas those values for kaolinite from Zebedielah varied fromh 15.6 to 17.7h and fromh −61 to −68 for δ18O and δD, respectively. The REE patterns and the contenth of otherh trace elementsh indicateh ongoing kaolinitisation in the Lwamondo and Zebediela kaolins with minimum mineral sorting. The sources of the kaolins varied from basic to acidic and these were derived from an active margin tectonic setting. Lwamondo kaolin was deposited in an oxic environment whereas Zebediela kaolin was deposited under suboxic/anoxic conditions. Based on the δ18O and δD values of the kaolinite, they formed in a supergene environment at temperatures generally below 40 ◦C.

Keywords: kaolinitisation; mineral sorting; rare earth elements; source rocks; supergene environment

1. Introduction Kaolin refers to a clayey rock and also to a group of clay minerals consisting of kaolinite, dickite, nacrite and halloysite. Kaolinite is the most abundant kaolin mineral, whereas dickite, nacrite and halloysite are relatively rare, being commonly formed by hydrothermal alteration and occurring in sedimentary and residual deposits in association with kaolinite [1]. Kaolin is used in many industrial applications, including paper ﬁlling and coating, refractory, ceramics, ﬁbreglass, cement, rubber, plastics, paint, catalyst and many other uses [2]. Considering its wide usage, any occurrence of kaolin is worth proper chemical, mineralogical and technological investigations [3]. A better understanding

Minerals 2019, 9, 93; doi:10.3390/min9020093 www.mdpi.com/journal/minerals Minerals 2019, 9, 93 2 of 19 of kaolins hinge on their environments of formation. Only a few studies have been carried out on the utilisation of kaolin in paleoenvironmental reconstruction. Trace element and stable isotope compositions of kaolins are powerful geochemical proxies in reconstructing paleoenvironments. Trace elements (rare earth elements (REEs), large ion lithophile elements, high-ﬁeld-strength elements and transition trace elements) are less mobile during sedimentary processes. This low mobility enables them to stay in the solid phase during erosion and sedimentation, thereby preserving the chemical signatures of their source rocks [4]. The concentrations of trace elements and speciﬁc ratios provide information on paleoweathering, sedimentary provenance, paleotectonics, hydraulic sorting, sediment recycling and paleo-redox conditions prevailing during the formation of kaolins [4–10]. The stable isotopic composition of kaolins can provide paleoenvironmental information on the isotopic composition of soil and meteoric waters, as well as paleotemperatures during the formation of the kaolins [11]. The stable isotope geochemistry of clay minerals has been applied in geothermometry, clay mineral genesis and diagenesis and paleoenvironments, amongst other disciplines [12–17]. In South Africa, there are a number of kaolin deposits distributed throughout the country. Most of the deposits have been studied for their geological, mineralogical and petrological properties. In Limpopo Province, the deposits have not been fully characterised, though some are being exploited for the making of bricks. This study, which is part of a mega-research on clays and clay minerals in Africa, focused on utilising trace elements and stable isotopes in two kaolins from the Lwamondo and Zebediela deposits in Limpopo Province, South Africa, to reconstruct the paleoenvironments in which they were formed.

2. Materials and Methods

2.1. Geologic Setting The Lwamondo kaolin covers an area of about 2700 m2 with an average thickness of 28 m. Half of the deposit has already been mined [18]. The Zebediela kaolin outcrop covers an area of about 3 km2, having an average thickness of 130 m and an estimated volume of 390,000,000 m3. Open pit mining is on-going, with only 40% of the deposit having been mined [18]. The Lwamondo and Zebediela kaolins are classiﬁed as secondary kaolins. The geology of the Lwamondo area is dominated by the Soutpansberg Group which is Proterozoic in age (Figure1). The rocks of the Soutpansberg Group overlie the eastern part of the Palala shear zone and part of the Kaapvaal Craton [19]. The Soutpansberg Group represents a volcano-sedimentary succession and is subdivided into seven formations: the Tshifhefhe, Sibasa, Funduzi, Willie’s Poort, Nzhelele, Stayt and Mabiligwe formations. Only four of the seven formations occur within the vicinity of the Lwamondo kaolin, namely the Tshifhefhe, Sibasa, Fundudzi and Willie’s Poort formations [18]. The basal discontinuous Tshifhefhe formation is a few meters thick and made up mainly of strongly epidotised clastic sediments which include shale, greywacke and conglomerate. The Sibasa formation is mainly a volcanic succession with rare discontinuous intercalations of clastic sediments, having a maximum thickness of about 3000 m [19]. The volcanics of the Sibasa formation comprise basalts, which were subaerially extruded and minor pyroclastic rocks [20]. The basalts are amygdaloidal, massive and generally epidotised. The clastic sediments, which include quartzite, shale and minor conglomerate, can locally have a maximum thickness of 400 m [19]. The overlying Fundudzi formation is only developed in the eastern part of the Soutpansberg Group and wedges out towards the west. Minerals 2019, 9, 93 3 of 19 Minerals 2019, 9, 93 3 of 20

Figure 1. Figure 1.Geological Geological setting setting of of the theLwamondo Lwamondo kaolinkaolin occurrence. Modified Modiﬁed from from [18]. [18 ].

TheThe Lwamondo Lwamondo kaolinkaolin was deposited deposited within within the theSibasa Sibasa formation formation volcanic volcanic succession. succession. The Thelithology lithology of ofthe the country country rocks rocks were were basalt, basalt, clastic clastic sediments sediments which which include include shale shale and and minor minor conglomeratesconglomerates and and minor minor pyroclastic pyroclastic rocks rocks with with a varicoloureda varicoloured kaolin kaolin (white, (white, brownish, brownish, red, red, pink pink and yellow).and yellow). The exposed The exposed outcrop outcrop at Lwamondo at Lwamondo had an had overburden an overburden with anwith average an average thickness thickness of 2–3 of m and2–3 consists m and ofconsists reddish of brownreddish soil. brown Kaolin soil. wasKaolin exposed was exposed in the lowerin the lower portion portion of the of outcrop. the outcrop. TheThe geology geology of of Zebediela Zebediela is is regionally regionally mademade up of rocks rocks that that belong belong to to the the Transvaal Transvaal Supergroup Supergroup (Figure(Figure2). 2). The The Transvaal Transvaal Supergroup Supergroup is is preserved preserved in in three three separateseparate basins—thebasins—the Transvaal, Transvaal, Kanye Kanye (Botswana)(Botswana) and and Griqualand Griqualand West West basins basins [21 ]—and[21]—and is subdivided is subdivided into into four four main main lithostratigraphic lithostratigraphic units, namelyunits, thenamely Protobasinal the Protobasinal rocks, Blackrocks, ReefBlack formation, Reef formation, Chuniespoort Chuniespoort Group Group and and Pretoria Pretoria Group Group [22 ]. The[22]. area The is coveredarea is covered by surﬁcial by surficial sediments sediments of the of Karoo the Karoo Supergroup Supergroup and and in thein the northern northern part part of of the Transvaalthe Transvaal Supergroup Supergroup is controlled is controlled by ENE- by and ENE- NNW-trending and NNW-trending faults, which faults, are which common are throughoutcommon thethroughout Kaapvaal Cratonthe Kaapvaal [18]. In Craton the Zebediela [18]. In the area, Zebediela the rocks area, belong the rocks to the belong Wolkberg to the GroupWolkberg followed Group by anfollowed unconformity-bound by an unconformity-bound Black Reef formation Black Reef (BRF), formation which (BRF), in turn which was overlain in turn was by theoverlain Chuniespoort by the Group.Chuniespoort The Chuniesport Group. The Group Chuniesport is of Proterozoic Group is age.of Proterozoic age. ThreeThree exposed exposed outcrops outcrops werewere studied and and these these were were named named as Quarry as Quarry One, One,Quarry Quarry Three and Three Quarry Four. An iron-rich vein, probably an extension of the Penge banded iron formation (BIF), cut and Quarry Four. An iron-rich vein, probably an extension of the Penge banded iron formation across Quarry Four. Above the iron-rich vein was a brownish to reddish kaolin layer. Kaolin of quarry (BIF), cut across Quarry Four. Above the iron-rich vein was a brownish to reddish kaolin layer. one was greyish and reddish in colour. Kaolin in Quarry three was whitish with no sedimentary Kaolin of quarry one was greyish and reddish in colour. Kaolin in Quarry three was whitish with no structure identified and within the quarry, the Zebediela kaolin was characterised by a major NNW- sedimentary structure identiﬁed and within the quarry, the Zebediela kaolin was characterised by a SE-trending fault truncated by numerous smaller veins. At Quarry Four, yellowish kaolin at the base majorwas NNW-SE-trending overlain by reddish fault brown, truncated pale brown by numerous and light smaller brown veins.kaolin Atlayers. Quarry These Four, colours yellowish suggested kaolin at thea high base iron was content overlain in the by deposit. reddish brown, pale brown and light brown kaolin layers. These colours suggested a high iron content in the deposit.

Minerals 2019, 9, 93 4 of 19 Minerals 2019, 9, 93 4 of 20

FigureFigure 2. Geological 2. Geological setting setting of of Zebediela Zebediela kaolinkaolin occurrence. occurrence. Modified Modiﬁed from from [18]. [ 18].

2.2. Sampling2.2. Sampling and Samples and Samples Nine samplesNine samples were were collected collected from from each each of of thethe twotwo study sites. sites. The The samples samples were were coded coded as as LWA1–LWA9 for the Lwamondo Kaolin and ZEB1–ZEB9 for the Zebediela Kaolin. The sampling LWA1–LWA9 for the Lwamondo Kaolin and ZEB1–ZEB9 for the Zebediela Kaolin. The sampling method was judgemental [3], that is, the number of samples and the sampling distance depended on methodthe was availability judgemental of expos [3ed], thatoutcrops is, the [23,24]. number Prior of to samplesanalyses, the and samples the sampling were pre-treated distance by depended sieving on the availabilityin a 63 µm of sieve, exposed removal outcrops of organic [23,24 matter,]. Prior disper to analyses,sion and theparticle samples size separation, were pre-treated based on by the sieving in a 63methodsµm sieve, described removal by [25] of organic and [26]. matter, Details dispersionof particle size and distribution particle size in the separation, studied kaolins based are on the methodsreported described in Table by [125 [27].,26]. The Details analyses of particle were perform size distributioned on the <2 inµm the fraction studied of the kaolins kaolin are samples, reported in Table1[which27]. Thecontain analyses more information were performed on their on environment the <2 µm of fraction formation. of the kaolin samples, which contain more information on their environment of formation. Table 1. Particle size distribution of the Lwamondo and Zebediela kaolins.

Table 1.SampleParticle size distributionSand (wt %) of the Lwamondo Silt (wt %) and Zebediela Clay (wt kaolins.%) LWA1 46 32 22 SampleLWA2 Sand (wt38 %) Silt (wt 40 %) Clay (wt 22 %) LWA1LWA3 4646 32 24 22 30 LWA2LWA4 3820 40 62 22 18 LWA3LWA5 4630 24 44 30 26 LWA4LWA6 2030 62 34 18 36 LWA5LWA7 3040 44 48 26 12 LWA6LWA8 3022 34 54 36 24 LWA7 40 48 12 LWA9 32 36 32 LWA8 22 54 24 ZEB1 24 32 44 LWA9 32 36 32 ZEB1ZEB2 2446 32 30 44 24 ZEB2ZEB3 4610 30 48 24 42 ZEB3ZEB4 1018 48 50 42 32 ZEB4 18 50 32 ZEB5 34 36 30 ZEB6 18 72 10 ZEB7 20 58 22 ZEB8 30 24 46 ZEB9 40 28 32 Minerals 2019, 9, 93 5 of 19

2.3. Laboratory Analyses Kaolin samples were collected from various sites within the Lwamondo and Zebediela study areas for trace element and stable isotope analyses. Trace element analyses were carried out at the Central Analytical Facility at Stellenbosch University, South Africa, while stable isotope analyses were done at the University of Cape Town. Laser Ablation Inductively Coupled Plasma Spectrometry (LA-ICP-MS) analysis was used to determine trace element concentrations in the samples using an Agilent 7700 instrument (Agilent Technologies, Santa Clara, CA, USA). A laser was used to vaporise the surface of the solid sample, whereas the vapour and any particles were transported by the carrier gas flow to the ICP-MS. The pressed pellet method for trace element analysis (8 ± 0.05 g) of milled powder was weighed and mixed thoroughly with three drops of Mowiol wax binder. The pellet was pressed with a pill press to 15 ton pressure and was dried in an oven at 100 ◦C for half an hour before being analysed. Ablation was performed in He at a flow rate of 0.35 L/min and then mixed with Ar (0.9 L/min) and N (0.004 L/min) just before introduction into the IC plasma. For traces in fusions, two spots of 100 µm were ablated on each sample using a frequency of 10 Hz and energy of 3.6 mJ/cm2. Trace elements were quantified using Basalt, Hawaiian Volcanic Observatory (BHVO) glass [28] and BHVO powder [29] as standards, employing standard–sample bracketing. Two replicate measurements were made on each sample. The calibration standard was run after every 12 samples. The instrument’s detection limits for the analysed trace elements are shown in Table2.

Table 2. Instrument detection limits (DL) of trace elements.

Trace Element DL (ppm) Trace Element DL (ppm) Sc 0.022 Pr 0.002 V 0.018 Nd 0.007 Cr 0.166 Sm 0.010 Co 0.008 Eu 0.002 Ni 0.144 Gd 0.007 Cu 0.020 Tb 0.001 Zn 0.130 Dy 0.004 Rb 0.008 Ho 0.001 Sr 0.003 Er 0.003 Y 0.002 Tm 0.001 Zr 0.005 Yb 0.007 Nb 0.001 Lu 0.001 Mo 0.009 Hf 0.004 Cs 0.003 Ta 0.001 Ba 0.020 Pb 0.005 La 0.001 Th 0.002 Ce 0.002 U 0.002

Oxygen isotope ratios of the <2 µm fraction of the kaolin samples were determined after being degassed under vacuum on the silicate line at 200 ◦C for two hours. The samples were reacted with ClF3 [30] in a conventional silicate line and the O2 was converted to CO2 using a hot platinised carbon rod. Hydrogen isotope analyses of absorbed water extracted in the manner described above were made using a variation of the closed tube Zn reduction method described by [31]. Hydrogen extraction for isotopic analysis followed the principle of [32]. After degassing at 180 ◦C in a vacuum to remove absorbed moisture, water was extracted from kaolin by heating in a Mo crucible with an induction furnace to >1500 ◦C.

3. Results The mineral phases and major oxides of the Lwamondo and Zebediela kaolins are reported in Reference [27]. Kaolinite was the most abundant mineral phase in both kaolin deposits, with means of 68.39 wt % and 77.73 wt % in the Lwamondo and Zebediela kaolins, respectively. Other mineral phases Minerals 2019, 9, 93 6 of 20

Minerals 2019, 9, 93 6 of 20 3. Results 3. ResultsThe mineral phases and major oxides of the Lwamondo and Zebediela kaolins are reported in ReferenceThe mineral [27]. Kaolinite phases andwas majorthe most oxides abundant of the Lwmineralamondo phase and in Zebediela both kaolin kaolins deposits, are reportedwith means in ReferenceofMinerals 68.392019 wt [27]. ,%9, 93 andKaolinite 77.73 wtwas % the in mostthe Lwamondo abundant mineraland Zebediela phase inkaolins, both kaolin respectively. deposits, Other with mineral means6 of 19 ofphases 68.39 thatwt % were and determined77.73 wt % ininclude the Lwamondo quartz and an smd ectiteZebediela in major kaolins, to minor respectively. quantities, Other clinochlore, mineral goethite and talc in minor to trace quantities and muscovite, microcline and plagioclase in trace phasesthat were that determined were determined include include quartz quartz and smectite and sm inectite major in to major minor to quantities, minor quantities, clinochlore, clinochlore, goethite quantities. The most abundant oxides were SiO2 and Al2O3, with respective means of 44.99 wt % and goethiteand talc and in minor talc in to minor trace quantitiesto trace quantities and muscovite, and muscovite, microcline microcline and plagioclase and plagioclase in trace quantities. in trace quantities.29.82 wt % The in the most Lwamondo abundant kaolin oxides and were 42.93 SiO wt2 and % andAl2O 31.763, with wt respective % in the Zebediela means of 44.99kaolin. wt % and The most abundant oxides were SiO2 and Al2O3, with respective means of 44.99 wt % and 29.82 wt % 29.82in the wt Lwamondo % in the Lwamondo kaolin and kaolin 42.93 wtand % 42.93 and 31.76wt % wtand % 31.76 in the wt Zebediela % in the kaolin.Zebediela kaolin. 3.1. Trace Elements 3.1.3.1. Trace TraceTrace Elements Elements element distribution in the clay fraction samples of the Lwamondo and Zebediela kaolins normalisedTraceTrace element element to upper distribution distribution continental in in crust the the clay clay(UCC) fraction fraction values samples samples are shown of of the the in Lwamondo LwamondoFigures 3 and and and 4, Zebediela Zebedielarespectively kaolins kaolins and normalisedthenormalised results areto to upper upperreported continental in Tables crust 3 and (UCC)(UCC) 4, respectively. valuesvalues are are shown shown in in Figures Figures3 and 3 and4, respectively 4, respectively and and the theresults results are are reported reported in Tablesin Tables3 and 3 and4, respectively. 4, respectively.

Figure 3. Upper continental crust (UCC) normalised trace element concentrations of the clay size FigurefractionFigure 3. 3.of Upper Upperthe Lwamondo continental continental kao crust crustlin (arranged (UCC) (UCC) normalisednormalised in increasing trac trace atomice element element number concentrations concentrations from Sc to U).of of the the clay clay size size fractionfraction of of the the Lwamondo Lwamondo kao kaolinlin (arranged (arranged in in increasing increasing atomic atomic number number from from Sc Sc to to U). U).

FigureFigure 4. 4. UCCUCC normalised normalised trace trace element element concentrations concentrations of of the the clay clay size size fraction fraction of of the the Zebediela Zebediela kaolin kaolin Figure(arranged(arranged 4. UCC in in increasing increasing normalised atomic atomic trace elementnumber number concentrationsfrom Sc to U). of the clay size fraction of the Zebediela kaolin (arranged in increasing atomic number from Sc to U).

Minerals 2019, 9, 93 7 of 19

Table 3. Trace element composition in ppm of the clay fraction of the Lwamondo kaolin (arranged in increasing atomic number from Sc to U).

Trace Elements and Selected Ratios LWA1 LWA2 LWA3 LWA4 LWA5 LWA6 LWA7 LWA8 LWA9 UCC 1 Sc 37.2 7.6 10.6 10.6 6.1 55.0 4. 9 85.5 3.7 14.0 V 230.4 28.1 26.6 49.6 12.7 82.7 8.6 738.0 6.5 97.0 Cr 85.6 75.0 171.6 145.5 168.9 3604.0 64.2 79.6 76.1 92.0 Co 15.1 1.8 4.9 3.8 1.5 48.6 1.4 42.7 1.3 17.3 Ni 178.8 123.3 93.0 122.0 269.0 1082.5 94.6 178.6 94.0 47.0 Cu 247.0 24.6 35.4.0 12.4 11.8 62.1 7.1 828.0 6.9 28.0 Zn 200.6 55.9 46 63.1 37.2 151.7 26.7 253.0 29.8 67.0 Rb 53.7 41.9 68.7 61.6 70.5 1.0 35.9 33.5 30.3 84.0 Sr 37.2 28.3 18.0 128.7 48.0 33.2 49.0 28.5 20.0 320.0 Y 17.4 2.4 3.4 19.4 2.2 15.5 1.9 52.4 1.2 21.0 Zr 140.1 21.1 13.3 45.0 1.8 29.0 99.5 266.4 59.0 193.0 Nb 12. 5 13.0 14.2 18.9 5.1 7.5 2.4 16.3 1.7 12.0 Mo 0.5 0.4 0.5 0.4 0.5 0.6 1.1 1.6 0.3 1.1 Cs 2.4 0.5 0.7 1.2 0.8 0.1 0.4 5.0 0.2 4.9 Ba 562.5 450.0 484.5 626.0 404.5 86.6 157.0 340.5 85.8 624.0 Hf 4.3 0.6 0.4 1.2 1.2 0.8 3.4 7.1 1.5 5.3 Ta 0.7 0.2 0.7 0.5 0.1 0.1 0.1 1.0 0.1 0.9 Pb 8.3 23.4 13. 5 30.1 9.4 9.8 35.8 11.2 11.4 17 Th 5.4 1.1 2.8 1.4 0.1 0.3 0.3 8.5 0.4 10.5 U 0.8 0.8 0.6 1.2 0.4 0.6 0.3 2.0 0.5 2.7 U/Th 0.2 0.7 0.2 0.8 3. 5 2.1 1.0 0.2 1.1 Ni/Co 11.9 67.4 19.2 32.1 181.8 22.3 70.1 4.2 74.0 V/Cr 2.7 0.4 0.2 0.3 0.1 0.0 0.1 9.3 0.1 1 Upper continental crust (UCC) values from [33]. Minerals 2019, 9, 93 8 of 19

Table 4. Trace element composition in ppm of the clay size fraction of the Zebediela kaolin (arranged in increasing atomic number from Sc to U).

Trace Elements and Selected Ratios ZEB1 ZEB2 ZEB3 ZEB4 ZEB5 ZEB6 ZEB7 ZEB8 ZEB9 UCC 1 Sc 20.9 30.6 83.6 33.3 62.3 46.4 30.6 63.7 41.8 14 V 61. 5 223.0 504.0 174.2 425.9 273.6 172.6 428.0 263.6 97 Cr 253.6 21.7 272.8 299.0 203.9 127.2 110.9 188.1 209.4 92 Co 92.3 16.1 4.5 6.2 33.8 43.8 32.4 22. 9 6.0 17.3 Ni 124.8 30.8 57.9 163.3 237.9 106.8 147.2 81.9 82.2 47 Cu 47.2 45.1 17.7 9.8 335.5 219.2 165.0 125.3 98. 5 28 Zn 133.4 51.7 12.9 13.9 80.2 26.8 130.8 57.3 37.0 67 Rb 6.8 115.2 120.1 10.3 14.6 13.3 57.3 37.2 17.8 84 Sr 2.9 23.5 6.6 1.8 1.7 2.0 0.9 2.0 2.2 320 Y 15.9 32.3 103.2 29. 5 14.5 14.1 12.2 13.2 11.3 21 Zr 47.4 172.2 232.9 257.0 150.2 63.8 148.4 217.6 147.6 193 Nb 10.5 10.8 3.3 9.1 3.08 11. 6 12.8 11.4 9.2 12 Mo 1.4 1.6 1.1 0.8 1.9 1.8 0.5 1.1 0.8 1.1 Cs 0.3 2.7 2.0 0.5 0.6 0.3 6.8 0.8 0.5 4.9 Ba 143.6 150.7 554.0 45.2 26.6 75.8 78.3 70.4 41.7 624 Hf 1.4 4.7 6.3 6.5 3.3 1.6 3.8 5.7 3.8 5.3 Ta 0.6 0.6 0.1 0.2 0.11 0.1 0.1 0.5 0.3 0.9 Pb 4.7 10.1 1.5 2.2 2.6 2.7 1.9 1.7 1.4 17 Th 7.6 18.7 3.5 4.8 1.5 0.8 0.5 1.6 0.9 10.5 U 1.3 2.4 9.0 5. 7 9.0 8.8 0.9 4.3 0.4 2.7 U/Th 0.2 0.1 2.6 1.18 6.0 11.5 1.8 2.8 0. 5 Ni/Co 1.4 1.9 12.9 26.21 7.0 2.4 4.5 3.6 13.8 V/Cr 0.2 10.3 1.9 0.58 2.1 2.2 1. 6 2.3 1.3 1 UCC values from [33]. Minerals 2019, 9, 93 9 of 19 Minerals 2019, 9, 93 9 of 20

Large ion lithophile elements (Rb, Ba, Sr, Th, U): ComparedCompared to thethe UCC,UCC, largelarge ionion lithophilelithophile elements are generallygenerally depleteddepleted in thethe LwamondoLwamondo kaolinskaolins (Figure(Figure3 3).).Rubidium Rubidium isis mostmost depleteddepleted inin LWA6, whereaswhereas Th Th is is more more depleted depleted in in LWA5. LWA5. Sample Sample LWA4 LWA4 has has almost almost the the same same concentration concentration of Baof asBa theas UCCthe UCC and and this samplethis sample is also is thealso least the depletedleast depleted in Sr. Thein Sr. UCC The normalised UCC normalised concentrations concentrations of large ionof large lithophile ion lithophile elements inelements the Zebediela in the kaolinsZebediela show kaolins that Sr show is the that most Sr depleted is the most trace depleted element and trace is depletedelement and in all is the depleted samples in (Figure all the4 ).samples Barium (Figure is slightly 4). depletedBarium is in slightly all Zebediela depleted samples. in all RubidiumZebediela issamples. only enriched Rubidium in ZEB2 is only and enriched ZEB3, Thin ZEB2 is only and enriched ZEB3, Th in is ZEB4 only and enriched U is depleted in ZEB4 and in samples U is depleted ZEB2, ZEB1,in samples ZEB5 ZEB2, and ZEB9. ZEB1, ZEB5 and ZEB9. High fieldfield strength elements (Zr, Hf, Y, Nb):Nb): Compared to the UCC, sample LWA8 is the only sample enriched in all high-field- high-field-strengthstrength elements (Figure 3 3).). NobiumNobium isis depleteddepleted in samples LWA5, LWA6, LWA7LWA7 and and LWA9, LWA9, whereas whereas it it is is slightly slightly enriched enriched in in samples samples LWA1, LWA1, LWA2, LWA2, LWA3, LWA3, LWA4 LWA4 and LWA8and LWA8 compared compared to the to UCC. the ZirconiumUCC. Zirconium is most is depleted most depleted in LWA5, in whileLWA5, Hf while is more Hf depleted is more indepleted LWA3. Inin theLWA3. Zebediela In the kaolins, Zebediela UCC kaolins, normalised UCC high-field-strength normalised high-field-strength elements are generally elements slightly are generally depleted, exceptslightly in depleted, ZEB2 and except ZEB4, in which ZEB2 areand similar ZEB4, towhich the UCC are similar (Figure to4). the UCC (Figure 4). Transition tracetrace elementselements (V,(V, Co,Co, Cu,Cu, NiNi andand Sc):Sc): NickelNickel isis enrichedenriched inin thethe LwamondoLwamondo kaolinskaolins compared to the the UCC UCC (Figure (Figure 3).3). Sc Sc is is enriched enriched in in samples samples LWA1, LWA1, LWA6 LWA6 and and LWA8 LWA8 and and depleted depleted in insamples samples LWA2, LWA2, LWA3, LWA3, LWA4, LWA4, LWA5, LWA5, LWA7 LWA7 and and LWA9 LWA9.. Vanadium Vanadium is only is only enriched enriched in LWA1 in LWA1 and andLWA8, LWA8, Co is Co enriched is enriched in LWA1 in LWA1 and andLWA6 LWA6 and andCu is Cu enriched is enriched in samples in samples LWA1, LWA1, LWA3, LWA3, LWA6 LWA6 and andLWA8 LWA8 (Figure (Figure 3). 3The). The UCC UCC normalised normalised transition transition trace trace elements elements are are generally enriched inin thethe Zebediela kaolins (Figure4 4).). ScandiumScandium isis enrichedenriched inin allall ZebedielaZebediela samples;samples; so is V, exceptexcept inin ZEB1.ZEB1. Co and Cu are slightlyslightly depleteddepleted inin ZEB3ZEB3 andand ZEB4.ZEB4. NiNi isis onlyonly depleteddepleted inin ZEB2.ZEB2.

3.2. Rare Earth Elements (REEs) Results ofof UCCUCC normalisednormalised REEsREEs ofof thethe Lwamondo Lwamondo and and Zebediela Zebediela kaolins kaolins are are plotted plotted in in Figures Figures5 and5 and6 and 6 and shown shown in Tables in Tables5 and 65. and There 6. isThere a slight is enrichmenta slight enrichment of heavy rareof heavy earth rare elements earth (HREEs) elements compared(HREEs) compared to light rare to light earth rare elements earth (LREEs)elements in (LREEs) these kaolins in these (Figures kaolins5 (Figuresand6). 5 and 6).

10.00 LWA1 LWA2 1.00 LWA3 LWA4 0.10 LWA5 Samples/UCC LWA6 0.01 LWA7 La Ce Pr NdSm Eu Gd Tb Dy Ho Er Tm Yb Lu LWA8 Rare earth elements

Figure 5. Rare earth element plot of the clay size fractionfraction of the Lwamondo kaolin samples compared toto thethe UCC.UCC.

Minerals 2019, 9, 93 10 of 19 Minerals 2019, 9, 93 10 of 20

10.00 ZEB1

ZEB2 1.00 ZEB3

ZEB4 0.10

Samples/UCC ZEB5 ZEB6 0.01 ZEB7 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Rare earth elements ZEB8

Figure 6. Rare earth elementelement plotplot ofofthe the clay clay size size fraction fraction of of the the Zebediela Zebediela kaolin kaolin samples samples compared compared to tothe the UCC. UCC.

TableTable 5. Rare earth element composition in ppm of thethe clay fraction of the Lwamondo kaolin.

1 RareRare Earth Elements LWA1 LWA2 LWA3 LWA4 LWA5 LWA6 LWA7 LWA8 LWA9 UCC earth LaLWA1 LWA2 10.1 LWA3 3.4 LWA4 6.6 38.6LWA5 2.7LWA6 12.4 LWA7 12.9 LWA8 30.8 2.5LWA9 31UCC 1 elements Ce 16.9 4.4 19.1 14.3 3.8 1.5 15.5 30.3 8.8 63 Pr 2.6 0.8 1.5 10.6 0.5 3.5 1.5 8.4 0.4 7.1 La 10.1 3.4 6.6 38.6 2.7 12.4 12.9 30.8 2.5 31 Nd 10.8 3.1 6.4 42.4 1.7 13.7 4.5 36.4 1.2 27 Ce Sm 16.9 4.4 2.7 19.1 0.8 1.514.3 10.23.8 0.31.5 3.715.5 0.8 9.430.3 0.28.8 4.7 63 Pr Eu 2.6 0.8 0.9 1.5 0.3 0.510.6 2.70.5 0.13.5 1.01.5 0.6 2.48.4 0.20.4 1 7.1 Nd Gd 10.8 3.1 2.7 6.40.6 0.842.4 6.61.7 0.513.7 3.2 4.5 0.5 8.636.4 0.21.2 4 27 Sm Tb 2.7 0.8 0.5 1.5 0.1 0.110.2 0.80.3 0.13.7 0.50.8 0.1 1.59.4 0.10.2 0.7 4.7 Dy 3.7 0.5 0.6 4.3 0.5 3.5 0.4 10.4 0.2 3.9 Eu Ho 0.9 0.3 0.8 0.5 0.1 0.22.7 0.70.1 0.21.0 0.60.6 0.1 2.22.4 0.10.2 83 1 Gd Er 2.7 0.6 2.5 0.8 0.3 0.46.6 1.80.5 0.33.2 1.80.5 0.3 6.68.6 0.20.2 2.3 4 Tb Tm 0.5 0.1 0.4 0.1 0 0.10.8 0.20.1 0.10.5 0.30.1 0.1 1.01.5 0.10.1 0.3 0.7 Dy Yb 3.7 0.5 2.5 0.6 0.3 0.44.3 1.20.5 0.33.5 1.70.4 0.3 7.110.4 0.20.2 2 3.9 Lu 0.4 0.1 0.1 0.2 0.1 0.2 0.1 1.0 0 0.3 Ho ΣLREE 0.8 0.1 46.7 13.40.2 36.40.7 125.40.2 9.60.6 390.1 36.3 126.32.2 13.50.1 137.883 Er ΣHREE 2.5 0.3 10.8 0.4 1.4 1.81.8 9.20.3 1.61.8 8.60.3 1.4 29.86.6 0.90.2 92.5 2.3 Tm ΣREE 0.4 0 57.5 14.8 0.1 38.30.2 134.60.1 11.20.3 47.6 0.1 37.8 156.11.0 14.40.1 230.30.3 Yb 2.5 0.3 0.4 1 1.2UCC values0.3 from [33]. 1.7 0.3 7.1 0.2 2 Lu 0.4 0.1 0.1 0.2 0.1 0.2 0.1 1.0 0 0.3 ΣLREETable 6.46.7Rare earth13.4 element 36.4 composition 125.4 in ppm9.6 of the clay39 size fraction36.3 of the126.3 Zebediela 13.5 kaolin. 137.8 ΣHREE 10.8 1.4 1.8 9.2 1.6 8.6 1.4 29.8 0.9 92.5 1 ΣREERare Earth Elements57.5 14.8 ZEB1 ZEB238.3 ZEB3134.6 ZEB411.2 ZEB547.6 ZEB6 37.8 ZEB7 ZEB8156.1 ZEB914.4 UCC230.3 La 33.1 51.01 4.4 UCC values 9.6 from 2.2 [33]. 2.6 3.5 2 2.5 31 Ce 41.1 129 18.6 64.3 12.5 21 25.4 7.4 2 63 TablePr 6. Rare earth element 7.9 10.1composition 2.7 in ppm 5.6 of the 0.8 clay size 1 fraction 1.2 of the 0.7Zebediela 0.8 kaolin. 7.1 Nd 31.1 36.6 13.6 24.9 3.9 5.9 5.8 2.8 3.8 27 Rare Sm 6.4 6.9 7 6.1 1.3 2.0 2.1 1.1 1 4.7 Eu 1.4 1.5 1.7 1.3 0.6 0.5 0.6 0.4 0.5 1 earth ZEB1 ZEB2 ZEB3 ZEB4 ZEB5 ZEB6 ZEB7 ZEB8 ZEB9 UCC 1 Gd 4.8 6.2 16 5.2 2.1 2.5 1.7 1.4 1.4 4 elements Tb 0.7 1 2.8 0.8 0.4 0.4 0.3 0.3 0.3 0.7 La Dy 33.1 51.0 3.9 6.14.4 179.6 5.12.2 2.72.6 2.93.5 2.4 2 2.5 1.92.5 3.931 Ce Ho 41.1 129 0.6 18.6 1.2 3.364.3 112.5 0.6 21 0.5 25.4 0.5 7.4 0.6 0.4 2 0.8 63 Pr Er 7.9 10.1 1.7 3.92.7 85.6 30.8 2 1 1.3 1.2 1.3 0.7 1.7 1.30.8 2.3 7.1 Tm 0.2 0.5 1.2 0.5 0.3 0.2 0.3 0.3 0.2 0.3 Nd Yb 31.1 36.6 1.3 13.6 3.5 6.424.9 3.5 3.9 2.85.9 1.25.8 1.8 2.8 2.5 1.83.8 2 27 Sm Lu 6.4 6.9 0.2 0.6 7 0.9 6.1 0.51.3 0.42.0 0.22.1 0.4 1.1 0.4 0.3 1 0.3 4.7 Eu ΣLREE 1.4 1.5 125.8 241.31.7 641.3 1170.6 23.40.5 35.50.6 40.3 0.4 15.8 120.5 137.8 1 Gd ΣHREE 4.8 6.2 8.6 16.8 16 39.6 5.2 14.42.1 9.22.5 6.71.7 71.4 8.3 6.21.4 10.3 4 ΣREE 134.4 258.1 103.6 131.4 32.6 42.2 47.3 24.1 18.2 148.1 Tb 0.7 1 2.8 0.8 0.4 0.4 0.3 0.3 0.3 0.7 1 Dy 3.9 6.1 17 UCC 5.1 values2.7 from [33].2.9 2.4 2.5 1.9 3.9 Ho 0.6 1.2 3.3 1 0.6 0.5 0.5 0.6 0.4 0.8 Er 1.7 3.9 8 3 2 1.3 1.3 1.7 1.3 2.3 Tm 0.2 0.5 1.2 0.5 0.3 0.2 0.3 0.3 0.2 0.3

Minerals 2019, 9, 93 11 of 20

Yb 1.3 3.5 6.4 3.5 2.8 1.2 1.8 2.5 1.8 2 Lu 0.2 0.6 0.9 0.5 0.4 0.2 0.4 0.4 0.3 0.3 ΣLREE 125.8 241.3 64 117 23.4 35.5 40.3 15.8 12 137.8 ΣHREE 8.6 16.8 39.6 14.4 9.2 6.7 7 8.3 6.2 10.3 ΣREE 134.4 258.1 103.6 131.4 32.6 42.2 47.3 24.1 18.2 148.1 Minerals 2019, 9, 93 1 UCC values from [33]. 11 of 19

3.3. Hydrogen and Oxygen Isotopes 3.3. Hydrogen and Oxygen Isotopes The stable isotope values of kaolinite from Lwamondo vary from 17.4‰ to 19.2‰ δ18O, with a 18 meanThe of 18.6‰ stable and isotope from values –54‰ of to kaolinite84‰ δD, fromwith a Lwamondo mean of −65.3‰ vary from (Table 17.4 7). Theto mean 19.2 waterδ O, content with a − − meanis 10.63%. of 18.6 The andZebediela from kaolinite54 to 84 has δtheD, withfollowing a mean stable of 65.3isotope(Table values:7).h Theδ18O mean valuesh water range content from 18 is16.7‰ 10.63%. to 17.7‰, Theh Zebediela with a mean kaoliniteh value hash of the16. following7‰ and the stable δD values isotopeh range values: betweenδ O values–68‰ rangeand –61‰, from − − 16.7with a tomean 17.7 value, with of – a65.3‰ mean value(Table of 7). 16. The 7 meanand water the δD content values in range the Zebediela between kaolins68 and is 10.86%.61 , − withThe hbinary a mean plot valueh of δ of18O and65.3 δD(Table values7 ).shows Theh meanthat most water samples content fall in in the the Zebediela supergene kaolins fieldh (Figure is 10.86%. h7). The binary plot of δ18O and hδD values shows that most samples fall in the supergene ﬁeld (Figure7).

Table 7. H2O yields and δ18O and δD values of the Lwamondo and Zebediela kaolins. 18 Table 7. H2O yields and δ O and δD values of the Lwamondo and Zebediela kaolins. Sample δ18O (‰) δD (‰) H2O (%) 18 Sample LWA1 δ O(17.4) −84δD( )H8.48 2O (%) h h LWA1LWA2 17.419.2 −58− 8412.05 8.48 − LWA2LWA9 19.219.1 −54 5811.37 12.05 LWA9 19.1 −54 11.37 MeanMean 18.618.6 −65− 6510.63 10.63 ZEB2ZEB2 16.716.7 −68− 6810.16 10.16 ZEB4ZEB4 17.717.7 −61− 6111.67 11.67 ZEB8ZEB8 15.615.6 −63− 6310.74 10.74 Mean 16.7 −64 10.86 Mean 16.7 −64 10.86

-20-100 10203040 60 40 20 0 -20 -40

D (‰) ZEB4 LWA2 δ -60 ZEB8 LWA9 -80 ZEB2 -100 LWA1 -120 -140 δ18O (‰)

Lwamondo kaolin Zebediela kaolin GMWL S/H Line Kaolinite Line

Figure 7.7. δ1818OO versus δD plot of clay fractions of thethe LwamondoLwamondo andand ZebedielaZebediela kaolins.kaolins. The global meteoric waterwater line line (GMWL), (GMWL), the the supergene–hypogene supergene–hypogene line (S/Hline (S/H line) line) and theand kaolinite the kaolinite line are line plotted are forplotted reference. for reference.

4. Discussion The distribution of trace elements in clays is inﬂuenced by sedimentary processes (such as weathering, sorting and sedimentary recycling) and sedimentary provenance, such as the nature of parent rocks [34,35]. The stable isotopic composition of kaolins contains information about temperatures. Minerals 2019, 9, 93 12 of 19

4.1. Paleoweathering The higher concentration of Cr and V in the studied kaolin may be related to their low mobility duringMinerals the kaolinitisation2019, 9, 93 process [36]. The Lwamondo and Zebediela kaolins were also enriched12 of 20 in Ni, except sample ZEB2 which is related to it being easily mobilized during weathering. Barium and Sr were4. Discussion depleted in all samples, indicating that they were easily mobilized during weathering and removedThe from distribution the environment of trace [elements37,38]. During in clays kaolinitisation, is influenced by Sc sedimentary was slightly processes concentrated (such as in the kaolinweathering, deposit and sorting was and enriched sedimentary in all samplesrecycling) except and sedimentary in samples provenance, LWA1, LWA6 such and as the LWA8. nature of Theparent HREEs rocks are[34,35]. more The enriched stable isotopic than LREEs comp inosition the weatheredof kaolins andcontains sub-weathered information zonesabout [39]. The REEtemperatures. patterns show more enrichment in HREEs than LREEs in all samples, with slight positive Eu anomaly due to the presence of plagioclase in the kaolin samples. The REE pattern and the content of other4.1. tracePaleoweathering elements show evidence of weathering related to kaolinitisation in the Lwamondo and ZebedielaThe kaolins. higher concentration of Cr and V in the studied kaolin may be related to their low mobility during the kaolinitisation process [36]. The Lwamondo and Zebediela kaolins were also enriched in 4.2. SedimentaryNi, except sample Provenance ZEB2 which and Tectonic is related Setting to it being easily mobilized during weathering. Barium and Sr were depleted in all samples, indicating that they were easily mobilized during weathering and The concentrations of some trace elements, such as Th, Sc, Cr, La, Ni and Co and their ratios varied removed from the environment [37,38]. During kaolinitisation, Sc was slightly concentrated in the in differentkaolin deposit source and rock was composition enriched in (silicic all sample ands basic). except Therefore,in samples LWA1, they are LWA6 widely and used LWA8. for provenance studies, asThe they HREEs are transported are more enriched almost than exclusively LREEs in inthe the weathered terrigenous and sub-weathered component of zones sediment, [39]. The thereby reflectingREE patterns the chemistry show more of their enrichment source rocksin HREEs [4]. than Felsic LREEs rocks in are all richersamples, in Thwith and slight La, positive whereas Eu mafic rocksanomaly are richer due in to Co, the Sc, presence Ni and of Cr plagioclase [9]. Hence, in th thesee kaolin elements samples. and The their REE ratios pattern (La/Sc, and the Th/Sc, content Th/Co and Cr/Th)of other trace could elements be used show to inferevidence the of sedimentary weathering related provenance to kaolinitisation or source in rockthe Lwamondo composition and [7,8]. FigureZebediela8 shows kaolins. a basic source of the Zebediela kaolins and an intermediate-to-silicic source of the Lwamondo kaolins, whereas Figure9 shows a felsic source of these kaolins. 4.2. Sedimentary Provenance and Tectonic Setting High concentrations of Cr (>150 ppm) and Ni (>100 ppm), low Cr/Ni ratios (<1.5) and a high correlationThe coefficient concentrations between of some Cr and trace Ni elements, (>0.90) are such indicative as Th, Sc, of Cr, an La, ultramafic Ni and Co rock and source, their ratios whereas a Cr/Nivaried ratio in >2.0 different indicates source mafic-volcanic rock composition detritus (silicic [40 and]. Concentrations basic). Therefore, of Crthey and are Ni widely in the used Lwamondo for and Zebedielaprovenance kaolinsstudies, wereas they very are varied,transported even almo amongst exclusively samples in in the the terrigenous same location. component Ingeneral, of sediment, thereby reflecting the chemistry of their source rocks [4]. Felsic rocks are richer in Th and Ni concentrations were greater than 100 ppm and Cr concentrations were greater than 150 ppm. La, whereas mafic rocks are richer in Co, Sc, Ni and Cr [9]. Hence, these elements and their ratios The correlation(La/Sc, Th/Sc, coefficient Th/Co and between Cr/Th) could Cr and be used Ni wasto infer 0.98 the and sedimentary 0.33 in the provenance Lwamondo or source and Zebediela rock kaolins,composition respectively. [7,8].This Figure shows 8 shows that a although basic source an ultramaficof the Zebediela or mafic kaolins source and could an intermediate-to- be inferred for the Lwamondosilicic source kaolin, of the such Lwamondo sources cannot kaolins, be whereas inferred Figure for the9 shows Zebediela a felsic kaolin. source of these kaolins.

10.00

1.00 Silicic rocks

0.10 Th/Co

0.01 Basic rocks

0.00 0.01 0.10 1.00 10.00 La/Sc Lwamondo Zebediela

FigureFigure 8. Th/Co 8. Th/Co vs. vs. La/Sc La/Sc showing showing thethe source rocks rocks of of the the Lwamondo Lwamondo and and Zebediela Zebediela kaolins. kaolins.

The Lwamondo kaolin deposit is located in the Sibasa formation and there are basalts which rest on the basements of the Hoot Plaats gneiss and granite. The observed kaolin deposit was ferruginous

Minerals 2019, 9, 93 13 of 19 and whitish. The geology of the area suggests that kaolin may have been formed from basalt and clastic sediments. The surrounding country rocks are believed to have provided the primary minerals for kaolinitisation. The ﬁeld observation from the Zebediela kaolin is similar to the one reported by [13,41] of the Kgwakgwe kaolins in Botswana. The thickness of the beds in the Zebediela kaolin deposit showed different colours (pinkish, yellowing and reddish) and layerings, which depict beds of a sedimentary nature. The kaolin deposit is penetrated by reddish veins which probably mark groundwater passages. Surrounding country rocks are arkose mudstone and shale which are considered to have provided primary minerals for kaolinitisation. The Th/U ratio can also give an indication of the source of kaolins [5]. Low Th/U ratios (<3.5) are usually found in mantle-derived rocks, though sediments from active margin tectonic settings with major components of young undifferentiated crust could also have Th/U ratios <3.5 [5]. Due to the low concentrations of U and Th in the Lwamondo and Zebediela kaolins (Figure 10), it is suggested that these kaolins are derived from an active margin tectonic setting. Minerals 2019, 9, 93 13 of 20

1.50

Basic rocks 1.00

Th/Sc Lwamondo 0.50 Felsic rocks Zebediela

0.00 123456789 Sample

MineralsFigure 2019Figure, 9,9. 93 9.Plot Plot of of Th/Sc Th/Sc ofof thethe LwamondoLwamondo and Zebe Zebedieladiela kaolins, kaolins, showing showing a asilicic silicic source. source. 14 of 20

High concentrations of Cr (>150 ppm) and Ni (>100 ppm), low Cr/Ni ratios (<1.5) and a high 8.00 ZEB2 correlation coefficient between Cr and Ni (>0.90) are indicative of an ultramafic rock source, whereas

a Cr/Ni ratio >2.0 indicates mafic-volcanic detritusLWA1 [40]. Concentrations of Cr and Ni in the Lwamondo 6.00and Zebediela kaolins were very varied, even among samples in the same location. In ZEB1 general, Ni concentrations were greater than 100 ppm and Cr concentrations were greater than 150 ppm. The correlation Uppercoefficient crust between Cr and Ni was 0.98 and Weathering0.33 in the trendLwamondo and LWA3 LWA8 Zebediela kaolins,4.00 respectively. This shows that although an ultramafic or mafic source could be Th/U inferred for the Lwamondo kaolin, such sources cannot be inferred for the Zebediela kaolin. The Lwamondo kaolin depositZEB9 is located in the Sibasa formation and there are basalts which rest 2.00 on the basements of the Hoot Plaats gneissLWA2 and granite. The observed kaolin deposit was ferruginous and whitish. The geologyLWA7 of theLWA9 area suggestsLWA4 that kaolin ZEB4may haveDepleted been formed mantle from basalt and LWA5 ZEB7 ZEB8 clastic sediments. The surroundingLWA6 countryZEB6 rocks are believedZEB3 to have provided the primary minerals 0.00 ZEB5 for kaolinitisation. 0.1 1 10 100 The field observation from the Zebediela kaolin is similar to the one reported by [13,41] of the Th Kgwakgwe kaolins in Botswana. The thickness of the beds in the Zebediela kaolin deposit showed different colours (pinkish, yellowing and reddish) and layerings, which depict beds of a sedimentary Lwamomdo Zebediela nature. The kaolin deposit is penetrated by reddish veins which probably mark groundwater passages. Surrounding country rocks are arkose mudstone and shale which are considered to have Figure 10. Plot of Th/U vs. Th for the Lwamondo and Zebediela kaolins, showing a mantle source. providedFigure primary 10. Plot ofminerals Th/U vs. for Th kaolinitisation. for the Lwamondo and Zebediela kaolins, showing a mantle source. 4.3.The Hydraulic Th/U ratio Sorting can and also Sediment give an Recycling indication of the source of kaolins [5]. Low Th/U ratios (<3.5) are usually found in mantle-derived rocks, though sediments from active margin tectonic settings with The Th/Sc versus Zr/Sc bivariate plot can be used to illustrate hydraulic sorting and sedimentary major components of young undifferentiated crust could also have Th/U ratios <3.5 [5]. Due to the recycling [5]. Whereas Th is incompatible in igneous systems, Sc is compatible in igneous systems [14]. low concentrations of U and Th in the Lwamondo and Zebediela kaolins (Figure 10), it is suggested Therefore, the Th/Sc ratio is a good proxy for igneous chemical differentiation processes. Unlike Zr, which that these kaolins are derived from an active margin tectonic setting. is strongly enriched in zircon, Sc is not enriched but generally preserves a signature of the provenance. The Zr/Sc ratio is a useful index for zircon enrichment [5]. Trend 1 on this plot (Figure 11) shows the normal igneous differentiation trend, which does not involve zircon enrichment. An enrichment of zircon occurs during sedimentary sorting or recycling (Trend 2) [10]. Figure 11 shows that the Lwamondo and Zebediela kaolins are generally around Trend 1, suggesting minimum mineral sorting [9].

Figure 11. Th/Sc vs. Zr/Sc plot showing the normal trend for compositional variation in the Lwamondo and Zebediela kaolins.

Minerals 2019, 9, 93 14 of 20

8.00 ZEB2

LWA1 6.00 ZEB1

Upper crust Weathering trend LWA3 LWA8 4.00 Th/U

ZEB9 2.00 LWA2 LWA7 LWA9 LWA4 ZEB4 Depleted mantle LWA5 ZEB7 ZEB8 LWA6 ZEB6 ZEB3 0.00 ZEB5 0.1 1 10 100 Th

Lwamomdo Zebediela

Minerals 2019, 9, 93 14 of 19 Figure 10. Plot of Th/U vs. Th for the Lwamondo and Zebediela kaolins, showing a mantle source.

4.3. Hydraulic Sorting and Sediment Recycling The Th/Sc versus versus Zr/Sc Zr/Sc bivariate bivariate plot plot can can be be used used to to illustrate hydraulic sorting and sedimentary recycling [5]. [5]. Whereas Whereas Th Th is incomp incompatibleatible in in igneous systems, Sc Sc is compatible in in igneous systems systems [14]. [14]. Therefore, the the Th/Sc Th/Sc ratio ratio is a is good a good proxy proxy for igneous for igneous chemical chemical differentiation differentiation processes. processes. Unlike UnlikeZr, which Zr, whichis strongly is strongly enriched enriched in zircon, in Sc zircon, is not enriched Sc is not but enriched generally but preserves generally a preservessignature aof signaturethe provenance. of the provenance.The Zr/Sc ratio The is a Zr/Scuseful ratioindex isfor a zircon useful enrichment index for zircon[5]. Trend enrichment 1 on this [plot5]. (Figure Trend 111) on shows this plotthe (Figurenormal 11igneous) shows differentiation the normal igneous trend, differentiationwhich does not trend, involve which zircon does enrichment. not involve zirconAn enrichment enrichment. of Anzircon enrichment occurs during of zircon sedimentary occurs during sorting sedimentary or recycling sorting (Trend or recycling2) [10]. Figure (Trend 11 2) shows [10]. Figure that the 11 Lwamondoshows that the and Lwamondo Zebediela and kaolins Zebediela are generally kaolins are around generally Trend around 1, suggesting Trend 1, suggesting minimum minimum mineral sortingmineral [9]. sorting [9].

Figure 11.11. Th/ScTh/Sc vs. vs. Zr/Sc Zr/Sc plot plot showing showing the the normal normal trend trend for for compositional compositional variation variation in the Lwamondo and Zebediela kaolins.

4.4. Paleo-Redox Conditions

U/Th ratios <1.25 suggest an oxic condition of deposition, whereas values >1.25 indicate suboxic and anoxic conditions [6]. Most of the Lwamondo kaolin showed low U/Th (0.15–3.45, average of 1.09), which indicates that the kaolin was deposited in an oxic environment; by contrast, the Zebediela kaolin displays high U/Th ratios (0.18–11.5, average of 2.95), which indicates that the kaolin was deposited under suboxic/anoxic conditions. A V/Cr ratio <2 indicates oxic, 2.0–4.25 indicates dysoxic and >4.25 indicates suboxic to anoxic conditions [6]. The V/Cr ratios of the Lwamondo kaolin samples varied from 0.02–9.27, with an average of 1.40, indicating an oxic condition (Figure 12), whereas the V/Cr ratios of the Zebediela kaolin varied from 0.24–10.28, with an average of 2.48, indicating an oxic condition (Figure 12). Minerals 2019, 9, 93 15 of 20

4.4. Paleo-Redox Conditions U/Th ratios <1.25 suggest an oxic condition of deposition, whereas values >1.25 indicate suboxic and anoxic conditions [6]. Most of the Lwamondo kaolin showed low U/Th (0.15–3.45, average of 1.09), which indicates that the kaolin was deposited in an oxic environment; by contrast, the Zebediela kaolin displays high U/Th ratios (0.18–11.5, average of 2.95), which indicates that the kaolin was deposited under suboxic/anoxic conditions. A V/Cr ratio <2 indicates oxic, 2.0–4.25 indicates dysoxic and >4.25 indicates suboxic to anoxic conditions [6]. The V/Cr ratios of the Lwamondo kaolin samples varied from 0.02–9.27, with an average of 1.40, indicating an oxic condition (Figure 12), whereas the MineralsV/Cr2019 ratios, 9, 93 of the Zebediela kaolin varied from 0.24–10.28, with an average of 2.48, indicating an oxic15 of 19 condition (Figure 12).

8 Suboxic/Anoxic

6 Oxic Dysoxic V/Cr

4 Dysoxic 2 Oxic 0 0 5 10 15 20 25 30 35 40 45 Ni/Co

Lwamondo Zebediela

FigureFigure 12. 12.Cross Cross plots plots of of V/Cr V/Cr vs.vs. Ni/Co used used as as paleo-redox paleo-redox proxies. proxies.

4.5. Paleotemperatures4.5. Paleotemperatures KaolinKaolin can can be be formed formed through through interactions interactions withwith modern local local meteoric meteoric water. water. The The isotopic isotopic compositioncomposition of kaoliniteof kaolinite was was determined determined usingusing thethe mean isotopic isotopic composition composition of ofmodern modern meteoric meteoric waterwater and and equilibrium equilibrium fractionation fractionation factors factors between between kaolinite kaolinite and and waterwater (Equations(Equations 1 (1) and and 2) (2))by [42]: by [42]: 𝛿O + 1000 𝛼 = (1) δ18O + 1000 O = 𝛿 OK + 1000 αK-W 18 (1) δ OW + 1000 𝛿D + 1000 𝛼 = (2) 𝛿D + 1000 D δDK + 1000 αK-W = (2) where 𝛼 and 𝛼 are equilibrium fractionδD Wfactors+ 1000 between kaolinite and water (meteoric water) with respect to oxygen and hydrogen [43]. where αO and αD are equilibrium fraction factors between kaolinite and water (meteoric water) KThe-W isotopicK-W composition of kaolinite was determined using the modern mean annual withtemperature respect to oxygen based on and Equations hydrogen 3 and [43 4:]. The isotopic composition of kaolinite was determined using the modern mean annual temperature 1000 ln 𝛼 = −2.2𝑥10𝑥 𝑇 −7.7 based on Equations (3) andHydrogen: (4): (3) Oxygen: 1000 ln 𝛼 = 2.76𝑥10 𝑥 𝑇 −6.75 (4) 6 −2 Hydrogen : 1000 lnαkaolinite−water = −2.2x10 x T − 7.7 (3) where T is the temperature (°K) and 𝛼 is equilibrium isotopic fractionation factors between kaolinite and water. Oxygen : 1000 lnα = 2.76x106x T−2 − 6.75 (4) The stable isotope composition of kaolins,kaolinite− aswater well as other clay minerals, is a function of the ◦ whereisotopicT is the composition temperature of ( K)the andwaterαkaolinite from− waterwhichis equilibriumthey were formed. isotopic fractionationThe equilibrium factors isotopic between kaolinite and water.

The stable isotope composition of kaolins, as well as other clay minerals, is a function of the isotopic composition of the water from which they were formed. The equilibrium isotopic fractionation factors between kaolinite and water were developed by [44] and [12]. These fractionation factors are a function of the temperature of kaolinitisation; therefore, the isotopic composition of kaolinite can provide information about its genesis [15]. The temperature of kaolinitisation (T) of the studied kaolins were calculated using Equation (5) [16] and is presented in Table8. The Lwamondo kaolin deposit had a mean temperature of 26.9 ± 3.6 ◦C whereas the Zebediela kaolin had a mean temperature of 36.6 ± 4.2 ◦C. Kaolinite in equilibrium with the global meteoric water line, kaolinitisation temperatures in the Lwamondo kaolins (26.9 ◦C), whereas the Zebediela kaolins kaolinitisation temperature was 36.6 ◦C. The Makoro and Kgwakgwe kaolins formed at around 40 ◦C[39]. Therefore, based on their stable isotopic compositions, the Lwamondo and Zebediela kaolins were formed at low temperatures (<40 ◦C), thus ruling out any Minerals 2019, 9, 93 16 of 19 possible hydrothermal process. The position of sample ZEB8 in the hypogene field could be interpreted as reflecting formation in isotopic equilibrium with its parental fluid without subsequent isotopic exchange with meteoric water [44].

6 −2 3.04x10 T = δOK − 0.125δDK + 7.04 (5)

Table 8. Calculated temperatures of kaolinitisation (T) of the analysed samples.

Samples δ18O( ) ∆d ( ) T (◦C) LWA1 17.4h −84h 21.8 LWA2 19.2 −58 28.1 LWA9 19.1 −54 30.9 MEAN 18.6 −65 26.9 ± 3.6 ZEB2 16.7 −68 33.9 ZEB4 17.7 −61 33.3 ZEB8 15.6 −63 42.5 MEAN 16.7 −64 36.6 ± 4.2

The mean temperature of kaolinitisation was determined using the global meteoric water line (GMWL). This was used to determine the mean stable isotopic composition of the meteoric water in equilibrium with the Lwamondo and Zebediela kaolins during their formation using Equations (3) and (4). The calculated mean isotopic composition of meteoric water is shown in Table9. Using the GMWL equation, the mean calculated isotopic composition of the meteoric water in 18 equilibrium with the Lwamondo kaolins is −5.3 and −33.2 for δ OW and δDW, respectively, whereas the mean calculated isotopic composition ofh meteoric waterh in equilibrium with the Zebediela 18 kaolin is −5.4 and −33.4 for δ OW and δDW, respectively (Table9 and Figure 13). The equilibrium fractionationh factor of deuteriumh (D) between kaolinite and water (αK-W D) of the Lwamondo and Zebediela kaolins calculated using the GMWL was 0.97 and 0.97, whereas the equilibrium fractionation factor of 18O between kaolinite and water of the Lwamondo and Zebediela kaolins was 1.024 and 1.022 (Table9).

Table 9. Mean isotopic composition of meteoric water and kaolinite–water fractionation factors.

18 ◦ 18 18 Samples δ OK ( ) δDK ( ) T ( K) δ OW ( ) δDW ( ) αK-W O αK-W D Lwamondo 18.6 h −65h 300.1 −5.3h −33.2h 1.024 0.97 ZebedielaMinerals 2019, 9, 93 16.7 −64 309.7 −5.4 −33.4 1.02217 0.97of 20

40 20 0 -20

D (‰) -40 δ -60 -80 -100 -15.0 -10.0 -5.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 δ18O (‰)

Lwamondo Zebediela GMWL S/H Line Kaolininte Line

Figure 13. δ18O vs. δD diagram for <2 µm fraction of the Lwamondo and Zebediela kaolins. Figure 13. δ18O vs. δD diagram for <2 µm fraction of the Lwamondo and Zebediela kaolins. The The GMWL, S/H Line and kaolinite lines are given as reference. GMWL, S/H Line and kaolinite lines are given as reference.

The isotopic compositions of the Lwamondo and Zebediela kaolins indicate that they are of weathering origin and formed in a supergene environment. Kaolins of weathering origin generally have higher δ18O (from 17‰ to 23‰) and δD (from –80‰ to 40‰) values. This is supported by data of the following kaolins of weathering origin: Nuevo Montecastelo kaolins in Spain [43], Variscan kaolins in Spain [16], Burela kaolin in Spain [45], Lastarria kaolin in Chile [17] and La Espingarda kaolin in Patagonia [46]. The major controlling factor in the clay mineral assemblages is the water composition in the environment of deposition [47]. The 2:1 layer silicates are dissolved by fresh water, which enhances kaolinite formation [48], whereas sea water preserves and promotes the genesis of mica, chlorite and smectite [49]. The mineralogy of the studied Lwamondo and Zebediela kaolins being dominated by kaolinite is indicative of relatively uniform chemistry from the processes of weathering, transportation, deposition and their reworkings.

5. Conclusions Trace element and stable isotope compositions of the Lwamondo and Zebediela kaolins were studied in order to determine the paleoenvironmental conditions underpinning their formation. The REE pattern and the content of various trace elements showed evidence of ongoing kaolinitisation in the Lwamondo and Zebediela kaolins, with minimum mineral sorting. The host rocks for the studied kaolins varied from basic to intermediate for the Zebediela kaolin deposit and from intermediate to felsic for the Lwamondo deposit. The U and Th contents of the Lwamondo and Zebediela kaolins suggest that these kaolins were derived from an active margin tectonic setting. The Lwamondo kaolin was deposited in an oxic environment, whereas the Zebediela kaolin was deposited under suboxic/anoxic conditions. The δ18O and δD values of the kaolinite from the Lwamondo and Zebediela kaolin deposits suggest that they were formed during weathering consistent with supergene origin. The mean δ18O and δD isotopic values of the studied samples indicate that the kaolin formation was in equilibrium with meteoric water at near-surface temperature. The temperature of formation of the Lwamondo and Zebediela kaolins based on their stable isotope values is indicative of the low temperature of kaolinitisation, ruling out any possible hydrothermal process.

Author Contributions: Conceptualization, Avhatakali Raphalalani and Georges-Ivo Ekosse; Formal analysis, Avhatakali Raphalalani; Funding acquisition, Avhatakali Raphalalani and Georges-Ivo Ekosse; Investigation, Avhatakali Raphalalani; Project administration, Georges-Ivo Ekosse; Supervision, Georges-Ivo Ekosse, John

Minerals 2019, 9, 93 17 of 19

The isotopic compositions of the Lwamondo and Zebediela kaolins indicate that they are of weathering origin and formed in a supergene environment. Kaolins of weathering origin generally have higher δ18O (from 17 to 23 ) and δD (from −80 to 40 ) values. This is supported by data of the following kaolins of weatheringh h origin: Nuevo Montecasteloh h kaolins in Spain [43], Variscan kaolins in Spain [16], Burela kaolin in Spain [45], Lastarria kaolin in Chile [17] and La Espingarda kaolin in Patagonia [46]. The major controlling factor in the clay mineral assemblages is the water composition in the environment of deposition [47]. The 2:1 layer silicates are dissolved by fresh water, which enhances kaolinite formation [48], whereas sea water preserves and promotes the genesis of mica, chlorite and smectite [49]. The mineralogy of the studied Lwamondo and Zebediela kaolins being dominated by kaolinite is indicative of relatively uniform chemistry from the processes of weathering, transportation, deposition and their reworkings.

Author Contributions: Conceptualization, A.R. and G.-I.E.; Formal analysis, A.R.; Funding acquisition, A.R. and G.-I.E.; Investigation, A.R.; Project administration, G.-I.E.; Supervision, G.-I.E., J.O. (John Odiyo), J.O. (Jason Ogola) and N.B.; Validation, G.-I.E.; Writing—original draft, A.R.; Writing—review and editing, G.-I.E., J.O. (John Odiyo), J.O. (Jason Ogola) and N.B. Funding: This research was funded by the National Research Foundation (South Africa), grant number UID 95472 and the University of Venda’s Research and Publications Committee grant number RI/14/RI/01. Acknowledgments: The authors are grateful to the management of Lwamondo Vhavenda Bricks and Zebediela Bricks for allowing the field component of the study to take place in their premises. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study, in the collection, analyses or interpretation of data, in the writing of the manuscript or in the decision to publish the results.

References

1. Murray, H.H. Applied Clay Mineralogy. Occurrences, processing and application of kaolins, bentonites, palygorskite-sepiolite and common clays. In Developments in Clay Science; Elsevier: Amsterdam, The Netherlands, 2007; Volume 2, p. 179. 2. Murray, H.H. Industrial Clays Case Study; No. 64; Mining, Minerals and Sustainable Development: London, UK, 2002. 3. Ekosse, G.E. Fourier transform infrared spectrophotometry and X-ray powder diffractometry as complementary techniques in characterizing clay size fraction of kaolin. J. Appl. Sci. Environ. Manag. 2005, 9, 43–48. [CrossRef] Minerals 2019, 9, 93 18 of 19

4. López, J.M.G.; Bauluz, B.; Fernández-Nieto, C.; Oliete, A.Y. Factors controlling the trace-element distribution in ﬁne-grained rocks: The Albian kaolinite-rich deposits of the Oliete Basin (NE Spain). Chem. Geol. 2005, 214, 1–19. [CrossRef] 5. McLennan, S.M.; Hemming, S.; McDaniel, D.K.; Hanson, G.N. Geochemical approaches to sedimentation, provenance, and tectonics. Geol. Soc. Am. 1993, 284, 21–40. 6. Jones, B.; Manning, D.A.C. Comparison of geochemical indices used for the interpretation of paleoredox condiions in ancient mudstones. Chem. Geol. 1984, 111, 111–129. [CrossRef] 7. Cullers, R.L. The controls on the major- and trace-element evolution of shales, siltstones and sandstones of Ordovician to Tertiary age in the Wet Mountains regions, Colorado, USA. Chem. Geol. 1995, 123, 107–131. [CrossRef] 8. Cullers, R.L. The geochemistry of shales, siltstones and sandstones of Pennsylvanian-Permian age, Colorado, USA: Implications for provenance and metamorphic studies. Lithos 2000, 51, 181–203. [CrossRef] 9. Armstrong-Altrin, J.S.; Nagarajan, R.; Madhavaraju, J.; Rosalez-Hoz, L.; Lee, Y.I.; Balaram, V.; Cruz-Martínez, A.; Avila-Ramírez, G. Geochemistry of the Jurassic and Upper Cretaceous shales from the Molango Region, Hidalgo, eastern Mexico: Implications for source-area weathering, provenance, and tectonic setting. C. R. Geosci. 2013, 345, 185–202. [CrossRef] 10. Nagarajan, R.; Armstrong-Altrin, J.S.; Kessler, F.L.; Hidalgo-Moral, E.L.; Dodge-Wan, D.; Taib, N.I. Provenance and tectonic setting of Miocene siliciclastic sediments, Sibuti formation, northwestern Borneo. Arab. J. Geosci. 2015, 8, 8549–8656. [CrossRef] 11. Sheldon, N.D.; Tabor, N.J. Quantitative paleoenvironmental and paleoclimatic reconstruction using paleosols. Earth-Sci. Rev. 2009, 95, 1–52. [CrossRef] 12. Sheppard, S.M.F.; Gilg, H.A. Stable isotope geochemistry of clay minerals. Clay Miner. 1996, 31, 1–24. [CrossRef] 13. Ekosse, G.E. Thermoanalytical characterization, stable isotopes and paleoenvironmental considerations of kaolinite from two genetic sources. Fresenius Environ. Bull. 2008, 17, 29–42. 14. Lee, Y.I. Geochemistry of shales of the Upper Cretaceous Hayang Group, SE Korea: Implications for provenance and source weathering at an active continental margin. Sediment. Geol. 2009, 215, 1–12. [CrossRef] 15. Fernández-Caliani, J.C.; Gálan, E.; Aparicio, P.; Miras, A.; Márquez, M.G. Origin and geochemical evolution of the Nuevo Montecastelo kaolin deposit (Galicia, NW Spain). Appl. Clay Sci. 2010, 49, 91–97. [CrossRef] 16. Clauer, N.; Fallick, A.E.; Gálan, E.; Aparicio, P.; Miras, A.; Fernández-Caliani, J.C.; Aubert, A. Stable isotope contraints on the origin of kaolin deposits from Variscan granitoids of Galicia (NW Spain). Chem. Geol. 2015, 417, 90–101. [CrossRef] 17. Gilg, H.A.; Hülmeyer, S.; Miller, H.; Sheppard, S.M.F. Supergene origin of the lastarria kaolin deposit, South-Central Chile, and paleoclimatic implications. Clays Clay Miner. 1999, 47, 201–211. [CrossRef] 18. Diko, M.L. Genesis and Ceramic Applications of Selected Kaolin Occurrences from Limpopo Province, South Africa and South West Region. Ph.D. Thesis, University of Limpopo, Polokwane, South Africa, 2011, unpublished. 19. Brandl, G.; Soutpansberg Group. Catalogue of South African Lithostratigraphic Units; SA Committee for Stratigraphy, Council Geoscience: Pretoria, South Africa, 2000; pp. 6-39–6-41. 20. Council for Geosciences. 1:250,000 Geological Series; Sheet 2228 Alldays; Council for Geosciences: Pretoria, South Africa, 2002. 21. Catuneanu, O.; Eriksson, P.J. The sequence stratigraphic concept and the Precambrian rock record: An example from the 2.7–2.1 Ga Transvaal Supergroup, Kaapvaal Craton. Precambrian Res. 1999, 97, 215–251. [CrossRef] 22. Eriksson, P.G.; Reczko, B.F.F. The sedimentary and tectonic setting of the Transvaal Supergroup ﬂoor rocks to the Bushveld Complex. J. Afr. Earth Sci. 1995, 21, 487–504. [CrossRef] 23. Ekosse, G. The Makoro kaolin deposit, southeastern Botswana: Its genesis and possible industrial applications. Appl. Clay Sci. 2000, 16, 301–320. [CrossRef] 24. Ekosse, G.E. Provenance of the Kgwakgwe kaolin deposit in southeastern Botswana and its possible utilization. Appl. Clay Sci. 2001, 20, 137–152. [CrossRef] 25. Jackson, M.L. (Ed.) Soil Chemical Analysis—Advanced Course, 2nd ed.; University of Wisconsin: Madison, WI, USA, 1979; p. 497. Minerals 2019, 9, 93 19 of 19

26. Van Reeuwijk, L.P. Procedures for Soil Analysis; Technical Paper 9; International Soil Reference and Information Centre: Wageningen, The Netherlands, 2002; p. 100. 27. Jochum, K.P.; Nohl, U.; Herwig, K.; Lammel, E.; Stoll, B.; Hofmann, A.W. GeoReM: A New Geochemical Database for Reference Materials and Isotopic Standards. Geostand. Geoanal. Res. 2005, 29, 333–338. [CrossRef] 28. Jochum, K.P.; Weis, U.; Schwager, B.; Stoll, B.; Wilson, S.A.; Haug, G.H.; Andreae, M.O.; Enzweiler, J. Reference Values Following ISO Guidelines for Frequently Requested Rock Reference Materials. Geostand. Geoanal. Res. 2016.[CrossRef]

29. Borthwick, J.; Harmon, R.S. A note regarding ClF3 as an alternative to BrF5 for oxygen isotope nalysis. Geochim. Cosmochim. Acta 1982, 46, 1666–1668. [CrossRef] 30. Diamond, R.E.; Harris, C. Oxygen and hydrogen isotope composition of Western Cape meteoric water. S. Afr. J. Sci. 1997, 93, 371–374. 31. Bigeleisen, J.; Perlman, M.L.; Prosser, H.C. Conversion of hydrogenic materials to hydrogen for isotopic analysis. Anal. Chem. 1952, 24, 1356–1357. [CrossRef] 32. Raphalalani, A. Paleoenvironmental Conditions Underpinning Kaolinitisation of Lwamondo and Zebediela Kaolin Deposits. Master’s Thesis, University of Venda, Thohoyandou, South Africa, 2017, unpublished. 33. Rudnick, R.L.; Gao, S. Composition of the continental crust. Treatise Geochem. 2003, 3, 1–64. 34. Nesbitt, H.W.; Markovics, G. Chemical processes affecting alkalis and alkaline earths during continental weathering. Geochim. Cosmochim. Acta 1980, 44, 1659–1666. [CrossRef] 35. Wronkiewicz, D.J.; Condie, K.C. Geochemistry of Archean shales from the Witwatersrand Supergroup, South Africa: Source-area weathering and provenance. Geochim. Cosmochim. Acta 1987, 51, 2401–2416. [CrossRef] 36. Vidal, P. Geochimie; Serie Geoscience 1998; Dunod: Paris, France, 1990. 37. Nyakairu, G.W.A.; Koebel, C.; Kurzweil, H. The Buwambo kaolin deposit in central Uganda: Mineralogical and chemical composition. Geochem. J. 2001, 35, 245–256. [CrossRef] 38. Mahjoor, A.S.; Karimi, M.; Rastegarlari, A. Mineralogical and Geochemical Characterisation of Clay Deposits from Soouth Abarkouh District of Clay Deposit (Central Iran) and their Applications. J. Appl. Sci. 2009, 9, 601–614. [CrossRef] 39. Zuoping, Z.; Chuanxian, L. The behaviour of rare-earth elements (REE) during weathering of granites in Southern Guangxi, China. Chin. J. Geochem. 1995, 15, 344–352. [CrossRef] 40. Garver, J.I.; Royce, P.R.; Smick, T.A. Chromium and nickel in shale of the Taconic foreland: A case study for the provenance of fine-grained sediments with an ultramafic source. J. Sediment. Res. 1996, 66, 100–106. 41. Ekosse, G.E. Kaolin deposits and occurrences in Africa: Geology, mineralogy and utilization. Appl. Clay Sci. 2010, 50, 212–236. [CrossRef] 42. Savin, S.M.; Epstein, S. Oxygen and hydrogen isotope geochemistry of clay minerals. Geochim. Cosmochim. Acta 1970, 4, 25–42. [CrossRef] 43. Lawrence, J.R.; Taylor, H.P., Jr. Deuterium and oxygen-18 correlation: Clay minerals and hydroxides in Quaternary soils compared to meteoric waters. Geochim. Cosmochim. Acta 1971, 35, 993–1003. [CrossRef] 44. Gilg, H.A.; Sheppard, S.M.F. Hydrogen isotope fractionation between kaolinite and water revised. Geochim. Cosmochim. Acta 1996, 60, 529–533. [CrossRef] 45. Gálan, E.; Aparicio, P.; Fernández-Caliani, J.C.; Miras, A.; Márquez, M.G.; Fallick, A.E.; Clauer, N. New insights on mineralogy and genesis of kaolin deposits: The Burela kaolin deposit (Northwestern Spain). Appl. Clay Sci. 2016, 131, 14–26. [CrossRef] 46. Dominguez, E.A.; Iglesias, C.; Dondi, M.; Murray, H. Genesis of the La Espingarda kaolin deposit in Patagonia. Appl. Clay Sci. 2010, 47, 290–302. [CrossRef] 47. Buhmann, C.; Buhmann, D. Hydrogen-isotope geochemistry of diagenetic clay minerals from Cretaceous Sandstones, Alberta, Canada: Evidence for exchange. Appl. Geochem. 1990, 5, 657–668. 48. Keller, W.D. Classification of kaolins exemplified by their textures in scan electron micrographs. Clays Clay Miner. 1978, 26, 1–20. [CrossRef] 49. Eberl, D.D. Clay minerals formation and transformation in rocks and soils. Philos. Trans. R. Soc. Lond. 1984, 311, 241–257. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).